INDEX

Explanations

references to instructions or tips

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ston

-0.16

eh

-0.15

 Sommer

-0.15

 Kane

-0.14

 Jensen

-0.14

Ð¾ÑĢÐ¾

-0.14

Ð¾ÐºÑĢÐµÐ¼Ð°

-0.13

esser

-0.13

lava

-0.13

dew

-0.13

POSITIVE LOGITS

0.15

indle

0.15

å²Ĺ

0.15

oui

0.15

kek

0.14

irket

0.14

XObject

0.14

 |--------------------------------------------------------------------------↵

0.14

ifetime

0.14

ÅĻez

0.14

Activations Density 0.066%

references to instructions or tips

No Comments

No Known Activations